cancel

Cancel any active OMC mode (autopilot, ralph, ultragoal, swarm, ultrapilot, pipeline, team) and clean up retired legacy state

Quality

69%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/cancel/SKILL.md

Quality

Content

77%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is an exceptionally actionable, well-sequenced operational runbook with genuine validation and failure handling for a destructive, multi-session operation. Its weaknesses are structural: the monolithic 370-line file inlines reference-grade material (fallback script, compatibility file list, worker cleanup) that belongs in one-level-deep reference files, and the cross-section repetition of the same safety rules inflates the token cost.

Suggestions

Move the emergency fallback script and the explicit legacy/global compatibility file list into reference files (e.g. references/emergency-fallback.md and references/legacy-compat.md) linked from short, clearly signaled sections, keeping SKILL.md as a lean overview of the four-phase flow.

Consolidate the lock/ownership and fail-closed rules, which are restated in "What It Does", "Cancellation Flow", "Authorized Global Pass", and "MCP Worker Cleanup", into a single stated-once contract section.

Extract the per-mode handling details (Team/Ralph linkage, Self-Improve worktree rules, MCP worker heartbeat handling) into a per-mode reference table or file so the main flow reads as one page.

Dimension	Reasoning	Score
Conciseness	Every token is domain-specific operational detail rather than concepts Claude already knows, but the same rules (lock/ownership enforcement, Team-first ordering, global-pass gating) are restated across "What It Does", "Cancellation Flow", "Preserved State", and "MCP Worker Cleanup" — noticeably tighter than ideal, though the repetition is arguably defensive emphasis for a destructive operation. Not 2 because there is no filler or explanation of known concepts; not 4 because the cross-section repetition could be consolidated.	3 / 5
Actionability	The body provides copy-paste-ready artifacts for every common case: the exact `ToolSearch` deferred-tool query, a complete bash argument parser, the exact `state_write`/`state_clear` call forms with parameters, the `cleanup-orphans.mjs` invocation with `--dry-run`, and a fully executable emergency fallback script (with its placeholders explicitly flagged).	5 / 5
Workflow Clarity	The four-phase sequence (Parse Arguments → Detect Active Modes → Cancel Each Session → Authorized Global Pass) has explicit validation checkpoints: fail-closed identity resolution, `state_get_status` before each mode cleanup, the per-mode dependency order, and explicit failure handling (retain state, report scope and reason, block the global pass) with retry-in-same-scope guidance — fully satisfying the destructive/batch-operation feedback-loop requirement.	5 / 5
Progressive Disclosure	There are no reference files at all: the ~90-line compatibility file list, the emergency fallback script, and the MCP worker cleanup details are all inlined in a ~370-line body. Section headers are clear and there is no nested-reference problem, so this is not a 2, but material that clearly belongs in separate reference files is inline with no external pointers, capping it at 3.	3 / 5
	Total	16 / 20 Passed

Description

61%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description states a concrete, well-scoped what with a distinctive enumerated mode list, but it omits any when-to-use guidance, which both caps completeness and weakens its trigger surface. Adding a "Use when..." clause naming the situations and user phrases that should invoke it would move it into the top band.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user asks to cancel, stop, or exit an active OMC mode (autopilot, ralph, swarm, etc.) or when a stop hook reports a completed mode needing cleanup."

Include natural synonyms users say, such as "stop" or "exit" (the body already documents `stopomc`), so the description's trigger coverage matches the body's.

Briefly state the outcome distinction (e.g. preserving resume state for some modes vs. clearing retired state) to make the capability set more comprehensive.

Dimension	Reasoning	Score
Specificity	"Cancel any active OMC mode" and "clean up retired legacy state" name the domain and two concrete actions, with the seven mode names enumerated, but coverage of what cancellation does per mode is not comprehensive — this sits between the 1-2-actions anchor (3) and several-actions-with-minor-gaps (4).	3 / 5
Completeness	The "what" is clear (cancel active modes, clean up retired legacy state) but there is no "Use when..." clause or equivalent trigger guidance, which the guidelines cap at 3. Not 2 because the what-half is concrete, not vague; not 4 because the when-half is entirely absent rather than merely implicit.	3 / 5
Trigger Term Quality	"Cancel" plus the enumerated mode names ("autopilot", "ralph", "ultragoal", "swarm", "ultrapilot", "pipeline", "team") are exactly the natural terms a user of this system would say, but common synonyms like "stop" or "exit" (which the body exposes as `stopomc`) are missing from the description.	4 / 5
Distinctiveness Conflict Risk	The enumerated OMC mode names carve a clear niche with minimal conflict risk against other skills; only generic "cancel that" requests unrelated to OMC modes could mis-trigger it, keeping it below the 5 anchor.	4 / 5
	Total	14 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: Yeachan-Heo/oh-my-claudecode
Path: skills/cancel/SKILL.md
Commit: dc7ba1d

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.